Project Info

GirlTalk AI

Devpost

This project did not submit a demo video on Devpost.

Inspiration

Our project was inspired by a real-life problem faced by one of our team members. Growing up as an only child with a strict mom, they felt like they didn't have a female role model to go to for advice. So they turned to online advice videos and podcasts from other women to provide that level of support. But sometimes, generalized videos don't cut it and you want advice that pertains to your specific situation. Enter GirlTalk AI.

What it does

GirlTalk AI allows you to write your concerns into our chatbox and receive a personalized podcast clip from our big sister host, Aisha. Our site also gives you some helpful resources from the web for anyone looking for mental health resources or helplines. You can also go back and listen to old podcast clips you've generated through our browse podcasts feature.

How we built it

Frontend: Lovable AI + React + TypeScript, styled with Tailwind CSS and shadcn/ui components. Podcast Generation: AI-powered script creation using the Hyperbolic API, with keyword-based matching for relevant topics. Data Storage: Podcast metadata and user info are stored using Firestore (Firebase). For this hackathon version, we use a constant user ID. UI: Multi-page Website with dedicated sections for Home (podcast summary and playback), Support (resource links), Browse (past episodes), and About (background info).

Challenges we ran into

Designing age-appropriate, empathetic AI responses. Managing data storage without user authentication. Connecting the TTS with our LLM.

Accomplishments we're proud of

Created a safe space for women around the world to express themselves and get the support they need. Built a fully functional and visually appealing website with personalized podcast generation. Integrated AI-generated scripts with real-time playback and resource curation. Had fun while doing it :)

What we learned

How to use multiple APIs in one project. How to integrate a database. The importance of thoughtful prompt engineering for sensitive topics. How to balance technical complexity with user safety and privacy. Effective use of modern frontend tools and cloud data storage.

What's next

Even more realistic TTS. Longer podcast episodes (up to an hour). Video component of Aisha "recording" the podcast episode for you.

Analysis

Compare with all teams

View

Metric

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only
  • CSSIn code
  • ExpressIn code
  • FirebaseIn code
  • HTMLIn code
  • JavaScriptIn code
  • PythonIn code
  • ReactIn code
  • Tailwind CSSIn code
  • TypeScriptIn code

9 of 9 appear in the indexed code.

AI coding agents

No AI coding agent signals were found in this repository.

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

272 KB

Source files

82

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.

0 stars